Math Professor’s Simulation Predicts CWS Winner

Monday, June 20 update:

Mathematics Professor Andrew Swift periodically runs his simulation through the College World Series, taking account of series wins and losses. Monday's new projection results follow.

Read through the original story to learn how Swift compiles the data and why more professional sports teams are hiring statisticians.

OMAHA - Good news Gators and Hurricanes fans, your teams have the best odds of winning this year’s College World Series, according to a University of Nebraska at Omaha (UNO) professor’s projection.

For the last few years, Andrew Swift, D.Sc., has crunched the numbers just before the series opener, determining the odds of each team becoming the next national champion.

Predicting the outcome of the greatest show on dirt all starts with ranking the teams by strength.

“We use all of the regular season games, to basically give you an idea of how good each team is,” Swift said.

Wins, losses and a few other variables go into the formula.

The result is a ranking of every Division I team. This year, Miami and Florida took the top spots. Every other team competing in the series is somewhere in the top 20.

The first part of the process is fairly painless, but it wasn’t always that way.

“Originally, I had to manually enter all of the data from the regular season,” Swift said. “The first time I did that, I think I worked on it for two days straight, just basically typing in all the data.”

Graduate student Andrew Tew had a solution.

“[He] wrote a little piece of code that basically goes on the web and sucks in all the data, what we call data-scraping,” Swift said. “We can grab the data almost instantaneously.”

Tew earned his master’s degree in 2015, but he still collaborates on projects like this one with Swift.

As soon as the superregionals finished this year, they pulled the data. Swift ranked the teams, then ran his projection shortly after the brackets were announced.

“I simulate exactly as it’s laid out with the double elimination,” he said. “I basically play it over and over again, 100,000 times and know what winning team comes out.”

The result is a table of numbers, reflecting how often each team won, or in other words, their odds of doing the same thing at T.D. Ameritrade Park.

Pulling for the Chanticleers? Swift says you shouldn’t give up hope.

“It’s such a small tournament that anything could really happen, as we know from previous years. That’s why they play the games.”

And those rooting for Miami or Florida shouldn’t feel too confident. The teams might be leading the simulation, but consider the statistics.

Swift said no matter who your favorite is, “it’s still more likely than not that one of the other seven teams is going to win.”

While Swift’s CWS project is just for fun, he said it highlights the growing role of statistics in professional sports.

“It’s grown from a pet project to teams hiring people to work on this stuff so they can get some kind of competitive edge.”

Odds are those staffers aren’t projecting tournament outcomes, but they may be trying to predict future performances of athletes, helping coaching staffs decide who to sign or who to trade for.

Every year, Swift is approached by students who are interested in math and sports, looking for a way to combine the two.

“I point them in the right direction,” he said.

Swift said it’s easy to direct them to good resources. There’s never been a better time to be interested in sports data and analysis.

“It used to be just maybe a dozen or so people who did it just for fun, […] and now you’ve got hundreds of people who are making livings off this and it’s just crazy.”

More Projections

Former graduate student Andrew Tew ran his own simulation. View Tew's results.

Want to read more about the CWS? Read about UNO's connections to the series or about the UNO College of Business Administration student set to sing the national anthem before Game Six.